The Golay Codes
Total Page:16
File Type:pdf, Size:1020Kb
Load more
Recommended publications
-
The Witt Designs, Golay Codes and Mathieu Groups
The Witt designs, Golay codes and Mathieu groups 1 The Golay codes Let V be a vector space over Fq with fixed basis e1, ..., en. A code C is a subset of V .A linear code is a subspace of V . The vector with all coordinates equal to zero (resp. one) will be denoted by 0 (resp. 1). The Hamming distance dH (u, v) between two vectors u, v ∈ V is the number P P of coordinates where they differ: when u = uiei, v = viei then dH (u, v) = |{i | ui 6= vi}|. The weight of a vector u is its number of nonzero coordinates, i.e., dH (u, 0). The minimum distance d(C) of a code C is min{dH (u, v) | u, v ∈ C, u 6= v}. The support of a vector is the set of coordinate positions where it has a nonzero coordinate. Theorem 1.1 There exist codes, unique up to isomorphism, with the indicated values of n, q, |C| and d(C): n q |C| d(C) name of C (i) 23 2 4096 7 binary Golay code (ii) 24 2 4096 8 extended binary Golay code (iii) 11 3 729 5 ternary Golay code (iv) 12 3 729 6 extended ternary Golay code Let us assume that the codes have been chosen such as to contain 0. Then each of these codes is linear. (The dimensions are 12, 12, 6, 6.) 1 The codes (i) and (iii) are perfect, i.e., the balls with radius 2 (d(C) − 1) around the code words partition the vector space. -
Geometry of the Mathieu Groups and Golay Codes
Proc. Indian Acad. Sci. (Math. Sci.), Vol. 98, Nos 2 and 3, December 1988, pp. 153-177. 9 Printed in India. Geometry of the Mathieu groups and Golay codes ERIC A LORD Centre for Theoretical Studies and Department of Applied Mathematics, Indian Institute of Science, Bangalore 560012, India MS received 11 January 1988 Abstraet. A brief review is given of the linear fractional subgroups of the Mathieu groups. The main part of the paper then deals with the proj~x-~tiveinterpretation of the Golay codes; these codes are shown to describe Coxeter's configuration in PG(5, 3) and Todd's configuration in PG(11, 2) when interpreted projectively. We obtain two twelve-dimensional representations of M24. One is obtained as the coUineation group that permutes the twelve special points in PC,(11, 2); the other arises by interpreting geometrically the automorphism group of the binary Golay code. Both representations are reducible to eleven-dimensional representations of M24. Keywords. Geometry; Mathieu groups; Golay codes; Coxeter's configuration; hemi- icosahedron; octastigms; dodecastigms. 1. Introduction We shall first review some of the known properties of the Mathieu groups and their linear fractional subgroups. This is mainly a brief resum6 of the first two of Conway's 'Three Lectures on Exceptional Groups' [3] and will serve to establish the notation and underlying concepts of the subsequent sections. The six points of the projective line PL(5) can be labelled by the marks of GF(5) together with a sixth symbol oo defined to be the inverse of 0. The symbol set is f~ = {0 1 2 3 4 oo}. -
Introduction to Coding Theory
Introduction Why Coding Theory ? Introduction Info Info Noise ! Info »?# Sink to Source Heh ! Sink Heh ! Coding Theory Encoder Channel Decoder Samuel J. Lomonaco, Jr. Dept. of Comp. Sci. & Electrical Engineering Noisy Channels University of Maryland Baltimore County Baltimore, MD 21250 • Computer communication line Email: [email protected] WebPage: http://www.csee.umbc.edu/~lomonaco • Computer memory • Space channel • Telephone communication • Teacher/student channel Error Detecting Codes Applied to Memories Redundancy Input Encode Storage Detect Output 1 1 1 1 0 0 0 0 1 1 Error 1 Erasure 1 1 31=16+5 Code 1 0 ? 0 0 0 0 EVN THOUG LTTRS AR MSSNG 1 16 Info Bits 1 1 1 0 5 Red. Bits 0 0 0 0 0 0 0 FRM TH WRDS N THS SNTNCE 1 1 1 1 0 0 0 0 1 1 1 Double 1 IT CN B NDRSTD 0 0 0 0 1 Error 1 Error 1 Erasure 1 1 Detecting 1 0 ? 0 0 0 0 1 1 1 1 Error Control Coding 0 0 1 1 0 0 31% Redundancy 1 1 1 1 Error Correcting Codes Applied to Memories Input Encode Storage Correct Output Space Channel 1 1 1 1 0 0 0 0 Mariner Space Probe – Years B.C. ( ≤ 1964) 1 1 Error 1 1 • 1 31=16+5 Code 1 0 1 0 0 0 0 Eb/N0 Pe 16 Info Bits 1 1 1 1 -3 0 5 Red. Bits 0 0 0 6.8 db 10 0 0 0 0 -5 1 1 1 1 9.8 db 10 0 0 0 0 1 1 1 Single 1 0 0 0 0 Error 1 1 • Mariner Space Probe – Years A.C. -
Reed-Solomon Encoding and Decoding
Bachelor's Thesis Degree Programme in Information Technology 2011 León van de Pavert REED-SOLOMON ENCODING AND DECODING A Visual Representation i Bachelor's Thesis | Abstract Turku University of Applied Sciences Degree Programme in Information Technology Spring 2011 | 37 pages Instructor: Hazem Al-Bermanei León van de Pavert REED-SOLOMON ENCODING AND DECODING The capacity of a binary channel is increased by adding extra bits to this data. This improves the quality of digital data. The process of adding redundant bits is known as channel encod- ing. In many situations, errors are not distributed at random but occur in bursts. For example, scratches, dust or fingerprints on a compact disc (CD) introduce errors on neighbouring data bits. Cross-interleaved Reed-Solomon codes (CIRC) are particularly well-suited for detection and correction of burst errors and erasures. Interleaving redistributes the data over many blocks of code. The double encoding has the first code declaring erasures. The second code corrects them. The purpose of this thesis is to present Reed-Solomon error correction codes in relation to burst errors. In particular, this thesis visualises the mechanism of cross-interleaving and its ability to allow for detection and correction of burst errors. KEYWORDS: Coding theory, Reed-Solomon code, burst errors, cross-interleaving, compact disc ii ACKNOWLEDGEMENTS It is a pleasure to thank those who supported me making this thesis possible. I am thankful to my supervisor, Hazem Al-Bermanei, whose intricate know- ledge of coding theory inspired me, and whose lectures, encouragement, and support enabled me to develop an understanding of this subject. -
The Binary Golay Code and the Leech Lattice
The binary Golay code and the Leech lattice Recall from previous talks: Def 1: (linear code) A code C over a field F is called linear if the code contains any linear combinations of its codewords A k-dimensional linear code of length n with minimal Hamming distance d is said to be an [n, k, d]-code. Why are linear codes interesting? ● Error-correcting codes have a wide range of applications in telecommunication. ● A field where transmissions are particularly important is space probes, due to a combination of a harsh environment and cost restrictions. ● Linear codes were used for space-probes because they allowed for just-in-time encoding, as memory was error-prone and heavy. Space-probe example The Hamming weight enumerator Def 2: (weight of a codeword) The weight w(u) of a codeword u is the number of its nonzero coordinates. Def 3: (Hamming weight enumerator) The Hamming weight enumerator of C is the polynomial: n n−i i W C (X ,Y )=∑ Ai X Y i=0 where Ai is the number of codeword of weight i. Example (Example 2.1, [8]) For the binary Hamming code of length 7 the weight enumerator is given by: 7 4 3 3 4 7 W H (X ,Y )= X +7 X Y +7 X Y +Y Dual and doubly even codes Def 4: (dual code) For a code C we define the dual code C˚ to be the linear code of codewords orthogonal to all of C. Def 5: (doubly even code) A binary code C is called doubly even if the weights of all its codewords are divisible by 4. -
Linear Codes and Syndrome Decoding
Linear Codes and Syndrome Decoding José Luis Gómez Pardo Departamento de Matemáticas, Universidade de Santiago de Compostela 15782 Santiago de Compostela, Spain e-mail: [email protected] Introduction This worksheet contains an implementation of the encoding and decoding algorithms associated to an error-correcting linear code. Such a code can be characterized by a generator matrix or by a parity- check matrix and we give procedures to compute a generator matrix from a spanning set of the code and to compute a parity-check matrix from a generator matrix. We introduce, as examples, the [7, 4, 2] binary Hamming code, the [24, 12, 8] and [23, 12, 7] binary Golay codes and the [12, 6, 6] and [11, 6, 5] ternary Golay codes. We give procedures to compute the minimum distance of a linear code and we use them with the Hamming and Golay codes. We show how to build the standard array and the syndrome array of a linear code and we give an implementation of syndrome decoding. In the last section, we simulate a noisy channel and use the Hamming and Golay codes to show how syndrome decoding allows error correction on text messages. The references [Hill], [Huffman-Pless], [Ling], [Roman] and [Trappe-Washington] provide detailed information about linear codes and, in particular, about syndrome decoding. Requirements and Copyright Requirements: Maple 16 or higher © (2019) José Luis Gómez Pardo Generator matrices and parity-check matrices Linear codes are vector spaces over finite fields. Maple has a package called LinearAlgebra:- Modular (a subpackage of the LinearAlgebra m (the integers modulo a positive integer m, henceforth denoted , whose elements we will denote by {0, 1, .. -
Cyclic Codes
Cyclic Codes Saravanan Vijayakumaran [email protected] Department of Electrical Engineering Indian Institute of Technology Bombay August 26, 2014 1 / 25 Cyclic Codes Definition A cyclic shift of a vector v0 v1 ··· vn−2 vn−1 is the vector vn−1 v0 v1 ··· vn−3 vn−2 . Definition An (n; k) linear block code C is a cyclic code if every cyclic shift of a codeword in C is also a codeword. Example Consider the (7; 4) code C with generator matrix 21 0 0 0 1 1 03 60 1 0 0 0 1 17 G = 6 7 40 0 1 0 1 1 15 0 0 0 1 1 0 1 2 / 25 Polynomial Representation of Vectors For every vector v = v0 v1 ··· vn−2 vn−1 there is a polynomial 2 n−1 v(X) = v0 + v1X + v2X + ··· + vn−1X Let v(i) be the vector resulting from i cyclic shifts on v (i) i−1 i n−1 v (X) = vn−i +vn−i+1X +···+vn−1X +v0X +···+vn−i−1X Example v = 1 0 0 1 1 0 1, v(X) = 1 + X 3 + X 4 + X 6 v(1) = 1 1 0 0 1 1 0, v(1)(X) = 1 + X + X 4 + X 5 v(2) = 0 1 1 0 0 1 1, v(2)(X) = X + X 2 + X 5 + X 6 3 / 25 Polynomial Representation of Vectors • Consider v(X) and v(1)(X) 2 n−1 v(X) = v0 + v1X + v2X + ··· + vn−1X (1) 2 3 n−2 v (X) = vn−1 + v0X + v1X + v2X + ··· + vn−2X h 2 n−2i = vn−1 + X v0 + v1X + v2X + ··· + vn−2X n h n−2 n−1i = vn−1(1 + X ) + X v0 + ··· + vn−2X + vn−1X n = vn−1(1 + X ) + Xv(X) • In general, v(X) and v(i)(X) are related by X i v(X) = v(i)(X) + q(X)(X n + 1) i−1 where q(X) = vn−i + vn−i+1X + ··· + vn−1X • v(i)(X) is the remainder when X i v(X) is divided by X n + 1 4 / 25 Hamming Code of Length 7 Codeword Code Polynomial 0000000 0 1000110 1 + X 4 + X 5 0100011 X + X 5 + X 6 1100101 1 + X + X 4 + X 6 0010111 X 2 + X 4 + X 5 + X 6 1010001 1 + X 2 + X 6 0110100 X + X 2 + X 4 1110010 1 + X + X 2 + X 5 0001101 X 3 + X 4 + X 6 1001011 1 + X 3 + X 5 + X 6 0101110 X + X 3 + X 4 + X 5 1101000 1 + X + X 3 0011010 X 2 + X 3 + X 5 1011100 1 + X 2 + X 3 + X 4 0111001 X + X 2 + X 3 + X 6 1111111 1 + X + X 2 + X 3 + X 4 + X 5 + X 6 5 / 25 Properties of Cyclic Codes (1) Theorem The nonzero code polynomial of minimum degree in a linear block code is unique. -
ABSTRACT IDEMPOTENTS in CYCLIC CODES by Benjamin Brame April 9, 2012
ABSTRACT IDEMPOTENTS IN CYCLIC CODES by Benjamin Brame April 9, 2012 Chair: Dr. Zachary Robinson Major Department: Mathematics Cyclic codes give us the most probable method by which we may detect and correct data transmission errors. These codes depend on the development of advanced math- ematical concepts. It is shown that cyclic codes, when viewed as vector subspaces of a vector space of some dimension n over some finite field F, can be approached as polynomials in a ring. This approach is made possible by the assumption that the set of codewords is invariant under cyclic shifts, which are linear transformations. Developing these codes seems to be equivalent to factoring the polynomial xn −x over F. Each factor then gives us a cyclic code of some dimension k over F. Constructing factorizations of xn − 1 is accomplished by using cyclotomic polyno- mials and idempotents of the code algebra. The use of these two concepts together allows us to find cyclic codes in Fn. Hence, the development of cyclic codes is a journey from codewords and codes to fields and rings and back to codes and codewords. IDEMPOTENTS IN CYCLIC CODES A Thesis Presented to The Faculty of the Department of Mathematics East Carolina University In Partial Fulfillment of the Requirements for the Degree Master of Arts in Mathematics by Benjamin Brame April 9, 2012 Copyright 2012, Benjamin Brame IDEMPOTENTS IN CYCLIC CODES by Benjamin Brame APPROVED BY: DIRECTOR OF THESIS: Dr. Zachary Robinson COMMITTEE MEMBER: Dr. Chris Jantzen COMMITTEE MEMBER: Dr. Heather Ries COMMITTEE MEMBER: Dr. David Pravica CHAIR OF THE DEPARTMENT OF MATHEMATICS: Dr. -
Arxiv:Math/0311319V1
Modular and p-adic cyclic codes* A. R. Calderbank and N. J. A. Sloane Mathematical Sciences Research Center AT&T Bell Laboratories Murray Hill, NJ 07974 ABSTRACT This paper presents some basic theorems giving the structure of cyclic codes of length n over the ring of integers modulo pa and over the p-adic numbers, where p is a prime not dividing n. An especially interesting example is the 2-adic cyclic code of length 7 with generator polynomial X3 + λX2 + (λ 1)X 1, where λ satisfies λ2 λ + 2 = 0. This is the 2-adic − − − generalization of both the binary Hamming code and the quaternary octacode (the latter being equivalent to the Nordstrom-Robinson code). Other examples include the 2-adic Golay code of length 24 and the 3-adic Golay code of length 12. 1. Introduction This paper was prompted by the following questions. It is known [14], [16] that the binary polynomial X3 +X +1 that generates the cyclic Hamming code of length 7 lifts to a polynomial 3 2 X + 2X + X + 3 over Z4 that generates the octacode, equivalent to the binary nonlinear Nordstrom-Robinson code. What codes are obtained if we continue to lift this polynomial to Z8, Z16,..., and even to the 2-adic integers Z2∞ ? What is the general structure of cyclic codes over these rings? (Sol´e[23] had already suggested in 1988 that p-adic cyclic codes should be investigated.) arXiv:math/0311319v1 [math.CO] 18 Nov 2003 The answer to the first question is given in Example 1 of Section 4, where we describe the “2-adic Hamming code” of length 7 in detail. -
Cyclic Codes, BCH Codes, RS Codes
Classical Channel Coding, II Hamming Codes, Cyclic Codes, BCH Codes, RS Codes John MacLaren Walsh, Ph.D. ECET 602, Winter Quarter, 2013 1 References • Error Control Coding, 2nd Ed. Shu Lin and Daniel J. Costello, Jr. Pearson Prentice Hall, 2004. • Error Control Systems for Digital Communication and Storage, Prentice Hall, S. B. Wicker, 1995. 2 Example of Syndrome Decoder { Hamming Codes Recall that a (n; k) Hamming code is a perfect code with a m × 2m − 1 parity check matrix whose columns are all 2m −1 of the non-zero binary words of length m. This yields an especially simple syndrome decoder look up table for arg min wt(e) (1) ejeHT =rHT In particular, one selects the error vector to be the jth column of the identity matrix, where j is the column of the parity check matrix which is equal to the computed syndome s = rHT . 3 Cyclic Codes Cyclic codes are a subclass of linear block codes over GF (q) which have the property that every cyclic shift of a codeword is itself also a codeword. The ith cyclic shift of a vector v = (v0; v1; : : : ; vn−1) is (i) the vector v = (vn−i; vn−i+1; : : : ; vn−1; v0; v1; : : : ; vn−i−1). If we think of the elements of the vector as n−1 coefficients in the polynomial v(x) = v0 + v1x + ··· + vn−1x we observe that i i i+1 n−1 n n+i−1 x v(x) = v0x + v1x + ··· + vn−i−1x + vn−ix + ··· + vn+i−1x i−1 i n−1 = vn−i + vn−i+1x + ··· + vn−1x + v0x + ··· + vn−i−1x n n i−1 n +vn−i(x − 1) + vn−i+1x(x − 1) + ··· + vn+i−1x (x − 1) Hence, i n i−1 i i+1 n−1 (i) x v(x)mod(x − 1) = vn−i + vn−i+1x + ··· + vn−1x + v0x + v1x + ··· + vn−i−1x = v (x) (2) That is, the ith cyclic shift can alternatively be thought of as the operation v(i)(x) = xiv(x)mod(xn − 1). -
A Combinatorial Construction of an M12-Invariant Code 1
Fifteenth International Workshop on Algebraic and Combinatorial Coding Theory June 18-24, 2016, Albena, Bulgaria pp. 325{329 A combinatorial construction of an 1 M12-invariant code Jurgen¨ Bierbrauer [email protected] Department of Mathematical Sciences Michigan Technological University Houghton, Michigan 49931 (USA) Stefano Marcugini, Fernanda Pambianco fstefano.marcugini, [email protected] Dipartimento di Matematica e Informatica, Universit`adegli Studi di Perugia, Via Vanvitelli 1, Perugia, 06123, Italy Abstract. In this work we summarized some recent results to be included in a forthcoming paper [2]. A ternary [66; 10; 36]3-code admitting the Mathieu group M12 as a group of automorphisms has recently been constructed by N. Pace, see [3]. We give a construction of the Pace code in terms of M12 as well as a combinatorial description in terms of the small Witt design, the Steiner system S(5; 6; 12): We also present a proof that the Pace code does indeed have minimum distance 36: 1 Introduction A large number of important mathematical objects are related to the Mathieu groups. It came as a surprise when N. Pace found yet another such exceptional object, a [66; 10; 36]3-code whose group of automorphisms is Z2 × M12 (see [3]). We present here two constructions for this code, an algebraic construction which starts from the group M12 in its natural action as a group of permutations on 12 letters, and a combinatorial construction in terms of the Witt design S(5; 6; 12): We also prove that the code has parameters as claimed. In the -
Finding Cyclic Redundancy Check Polynomials for Multilevel Systems James A
View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by University of Richmond University of Richmond UR Scholarship Repository Math and Computer Science Faculty Publications Math and Computer Science 10-1998 Finding Cyclic Redundancy Check Polynomials for Multilevel Systems James A. Davis University of Richmond, [email protected] Miranda Mowbray Simon Crouch Follow this and additional works at: http://scholarship.richmond.edu/mathcs-faculty-publications Part of the Mathematics Commons Recommended Citation Davis, James A., Miranda Mowbray, and Simon Crouch. "Finding Cyclic Redundancy Check Polynomials for Multilevel Systems." IEEE Transactions on Communications 46, no. 10 (October 1998): 1250-253. This Article is brought to you for free and open access by the Math and Computer Science at UR Scholarship Repository. It has been accepted for inclusion in Math and Computer Science Faculty Publications by an authorized administrator of UR Scholarship Repository. For more information, please contact [email protected]. 1250 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 46, NO. 10, OCTOBER 1998 Finding Cyclic Redundancy Check Polynomials for ,Multilevel Systems James A. Davis, Miranda Mowbray, and Simon Crouch Abstract-This letter describes a technique for finding cyclic for using all nine levels rather than, effectively, just eight, redundancy check polynomials for systems for transmission over in that with nine levels there are more possible choices for symmetric channels which encode information in multiple voltage code words which do not encode data, but are special signals levels, so that the resulting redundancy check gives good error protection and is efficient to implement. The codes which we used for synchronization information or to signal the end of construct have a Hamming distance of 3 or 4.